Return Path Signals Notes

Return Path Signals

Digital Modulation

• Digital modulation of analog RF carriers transports data from customer premises over the return path of HFC networks to the headend.
• A digital signal represents information as a succession of distinct ones and zeros.
  • One: Presence of electricity (or higher voltage).
  • Zero: Absence of electricity (or low voltage).

Binary Bits

• Each distinct one or zero is a binary bit, the smallest unit of data.
• Digital signals are modulated onto analog RF carriers by altering the carrier's:
  • Amplitude
  • Frequency
  • Phase
  • Combinations of all three
• Digital modulation makes reception and decoding possible over noisy networks.

Trade-offs in Digital Modulation

• Numerous digital modulation formats exist, offering trade-offs between:
  • Reliable reception/decoding.
  • Data throughput (amount of data transported over a given period).
• High data throughput often comes at the expense of reliable transport in noisy networks.

Elements of RF Carrier Modulation

• Three elements of an RF carrier can be modulated:
  • Amplitude
  • Frequency
  • Phase
• Advanced modulation schemes use combinations for higher throughput in noisier environments.

Types of Digital Modulation

Amplitude Shift Keying (ASK)

• RF carrier is a series of pulses coinciding with digital bits.
  • Carrier turned OFF for a zero bit.
  • Carrier turned ON for three cycles for a one bit.
• ASK decoder converts on/off signals back to data pulses.
• Susceptible to noise, which can cause errors (interpreting noise as a one bit).

Frequency Shift Keying (FSK)

• First digital modulation used for return paths.
• Frequency of the RF carrier is shifted to represent one or zero bits.
• FSK decoder senses the presence of the carrier and frequency variations.
• FSK modulation ignores amplitude variations, providing reliable transport despite noise.

Phase Shift Keying (PSK)

• Provides reliable delivery over noisy networks but with low data throughput.
• The phase of the RF carrier changes $180^\circ$ with every transition between a one and a zero bit.
• Requires a start/timing point to distinguish changes.
• PSK modulation ignores amplitude variations, enabling reliable data transport.
• FSK and PSK modulations are unsuitable for high data throughput applications like cable modems or EMTAs (Embedded Multimedia Terminal Adapters).

Quadrature Phase Shift Keying (QPSK)

• Transports data at higher throughput and with less bandwidth than FSK and PSK.
• Delivers two bits of data at one time.
• Used in cable modems, EMTAs, and set-top boxes (STB).
• Data stream is grouped into pairs of bits, called symbols (one or more data bits relating to a single modulation state transition) that represent four different number states: 00, 01, 10, and 11.
• Symbols are divided into two signals, with one signal's phase shifted $90^\circ$ from the other.
• The two signals are multiplexed before being modulated onto an RF carrier.
• Multiplexing places each symbol within one of the four quadrants.
• If a transported symbol falls within the decision boundary of its quadrant, it is decoded as valid.

Quadrature Amplitude Modulation (QAM)

• QPSK is robust enough for most RF return applications over an HFC network, although less robust than FSK or PSK in noisy environments.
• Operators are reducing noise to enable reliable reception/decoding of QAM schemes, thus implementing more symbols and bits for higher throughput rates.
• QAM and QPSK are similar because the data symbols in both schemes are placed into quadrants. For this reason, some people refer to QPSK as 4-QAM
• However, QPSK uses PSK to modulate the RF carrier, while QAM uses amplitude and phase modulation.

16-QAM

• Incoming data stream divided into groups of four-bit symbols.
  • Every clock cycle: the time that elapses from one read or write operation to another in the central processing unit of a computer or comparable device.
• The first two bits are routed to an amplitude modulator, and the other two to a second modulator.
• Both modulators share the same local oscillator; phase of the second modulator's signal is shifted by $90^\circ$.
• Modulator outputs are multiplexed into one channel with four bits of data per clock cycle.
• Multiplexing creates a carrier that is effectively both phase and amplitude modulated.
• 16-QAM creates 16 different symbol representations.
• 16-QAM symbols land on a constellation grid.
• QPSK creates only four symbols.
• Throughput rate of 16-QAM is four times that of QPSK.
• The name for a QAM modulation indicates the number of symbol combinations.
• Higher throughput comes at the expense of reliable reception in noisy environments because decision boundaries are reduced in size.
  • The Quadrature Amplitude Modulation (QAM) uses amplitude and phase modulation to modulate the RF carrier.
• Bit rates are typically expressed as the number of binary bits per second (bps) that are transported through the network.

Bit Rate and Symbol Rate

• In QPSK and QAM, bits are grouped into symbols processed during one clock cycle.
• Each dot on the bipolar graph represents one symbol.
• The number of bits in a symbol relates directly to the number of symbols created.
• Bit rate: Bits per second (bps).
• Symbol rate: Symbols per second (sym/s).
• QPSK symbols have two bits, so the bit rate is equal to the symbol rate multiplied by two.
• 16-QAM's bit rate is equal to the symbol rate multiplied by four (bits per symbol).
• Theoretically, symbol rate equals carrier bandwidth (e.g., 1 MHz RF carrier = 1 Msym/s for a bit rate of 2 Mbps).
• Filtering suppresses unnecessary sidebands and harmonics that interfere with other channels.
• Filters have a specified roll-off factor, reducing the symbol rate to less than the RF bandwidth of the carrier.
• DOCSIS calls for a roll-off factor of 25%, making channel bandwidth 125% more than the symbol.
• Symbol rate within a 3.2 MHz QPSK carrier is 2.56 Msym/s (3.2 MHz ÷ 1.25 = 2.56 Msym/s).

QPSK Carrier Bandwidth Examples

16-QAM Carrier Bandwidth Examples

64-QAM Carrier Bandwidth Examples

Modulation Scheme and Network Quality

• Decoding high bit rate modulation schemes on wide bandwidth carriers requires less noise and interference.
• The modulation scheme and return carrier bandwidth depend on transport network quality and traffic volume.
• Improvements allow reliable transport of 64-QAM on carriers up to 6.4 MHz.

Orthogonal Frequency Division Multiplexing (OFDM)

• Digital Signal Processing (DSP): Mathematical manipulation of digital data for fast operations.
• DSP enables practical applications of OFDM.
• OFDM is in the DOCSIS 3.1 specification and wireless technologies.
• Legacy modulation schemes modulate data onto a single carrier.
• Filters suppress harmonics and sidebands but reduce effective bandwidth.
• OFDM subcarriers' orthogonal spacing eliminates the need for filters and guard bands.
• OFDM channels can be modulated at the maximum order of modulation suitable for the network condition.
• OFDM approaches the theoretical maximum spectral efficiency for its bandwidth.

RF Carrier Elements (Recap)

• Amplitude, frequency, and phase can be modulated for data transport.
• QPSK groups data into symbols representing four number states (00, 01, 10, 11).
• Higher-order QAM offers higher data rates but is more vulnerable to noise.
• Filters require additional RF bandwidth due to roll-off factor.
• OFDM uses bandwidth and frequency spectrum more efficiently than legacy SC-QAM channels.

Spectrum Sharing Technologies

• Limited frequency spectrum requires transmission devices to share an RF carrier or channel.
• Access control protocols manage transmission from devices like cable modems, EMTAs, STBs, and cable gateways.
• Protocols help recognize signals transported through noisy networks.
• Return path technicians should recognize characteristics of different access control protocols.

Shared Media Access Control (MAC)

• Ensures smooth data traffic flow and prevents collisions.
• Ensures each transmission device has fair access to the return frequency band and correct data throughput.
• Allows the return path channel to be used at its maximum capacity.

MAC Protocols

• Various MAC protocols are used in different applications, appropriate for their transport media.
• Ethernet uses carrier sense multiple access with collision detection (CSMA/CD).
  • Devices listen for transmissions before beginning their transmission.
  • Devices suspend transmission if transmission from other devices is detected and attempts a retransmission after a prescribed delay.
• Carrier sense multiple access with collision avoidance (CSMA/CA): Requires the transmitter to receive acknowledgement that a packet was received before sending more packets.
  • When a collision occurs and is detected, the MAC instructs the sending devices to transmit again after a random amount of delay.

Real-Time Applications

• In high-traffic networks, the retransmission of data becomes inefficient as traffic congestion impacts real-time applications:
  • An application that specifies latency to be less than a defined value for the user to perceive the application as functioning with immediate response.
  • A telephone conversation is an example of a real-time application.
• Wi-Fi uses CSMA/CA: Sending devices listen for network traffic before transmitting.

ALOHA and TDMA

• There are two distinctly different MAC protocols specified for digital set-top boxes (STB); ALOHA and time division multiple access (TDMA).

ALOHA

• Developed by the University of Hawaii for channel allocation in ground-based radio communications, ALOHA is a collision recovery protocol.
  • If a collision occurs, the random waiting period for the next transmission varies between transmitters so the next transmission attempt will be successful.
• Before sending more data, the sending devices wait for acknowledgement of the packets receipt.
  • Without receipt of an acknowledgement, each device has a random waiting period before sending the data again.
  • Because the data is sent at different times, successful reception is likely.
  • Multiple data packets (A1, B2, C1, D1, D2, and E2) coming from multiple sources on the same carrier frequency without any separation in time between the packets. Without an interval in time between packets, the receiver interprets the information in the packets as garbled data, and all are rejected.
• Allows quick communication between STB and headend controller (HC), making it ideal for interactive products like video on demand (VOD).
• A customer can play, pause, or resume a VOD program with minimal delay.
• Retrying a command usually works without perception of a problem.

Time Division Multiple Access (TDMA)

• Time Division Multiple Access: A multiplexing scheme that allows two or more signals to be transported on the same frequency or channel using a time-sharing technique in which each signal occupies the channel for only a fraction of the time to dedicated receivers.
  • A subset of TDM, the access portion of TDMA, refers to the multiplexing technique that grants multiple devices access to the shared network and enables them to send data in an orderly fashion and still be received.
• Each STB is provided a time slot in which it is the only time the STB can send data, making for very reliable and orderly transport and reception of data.
• STB must request a time slot from the STB headend controller before sending data.
• The channel-based MAC protocol of TDMA is more reliable than ALOHA, but at slower response.
• Data is concentrated into packets and sent out when allocated by the MAC.
• To prevent packet delay in real-time applications, the MAC (1.1 and higher of the DOCSIS) assigns priority tags to each packet which are used to shape, police, and prioritize Internet protocol (IP) traffic according to the application's QoS traffic parameters.
• Network routers prioritize real-time data packets ahead of other traffic (email, etc.).
• TDMA: Each STB is provided a time slot to send data, enabling more reliable access than ALOHA.
• Time Division Multiple Access: A multiplexing scheme that allows two or more signals to be transported on the same frequency or channel using a time-sharing technique in which each signal occupies the channel for only a fraction of the time to dedicated receivers.

Achieving Higher Data Throughput

• Achieving Higher Data Throughput: Higher data rates in the return path can be achieved by transporting the data using higher orders of modulation and wider bandwidths.
• Advanced Physical Layer (PHY) technologies were developed and added to the DOCSIS 2.0 specification

Advanced Time Division Multiple Access (A-TDMA)

• Advanced time division multiple access (A-TDMA) - A time division multiple access (TDM) protocol with enhancements that include improved forward error correction (FEC) and ingress cancelation.

Synchronous Code Division Multiple Access (S-CDMA)

• Synchronous code division multiple access (S-CDMA) - A coded network access technique in which the data is coded and spread across the spectrum using an algorithm that places the data in an orthogonal pattern that enables the receiver to identify and decode data that is received simultaneously.
• Maximum bit rates of A-TDMA and S-CDMA are the same.
• A-TDMA is easier to transition into from early DOCSIS versions.
• A-TDMA is more widely adopted than S-CDMA, even though it may perform better than A-TDMA under certain operating conditions.

DOCSIS Parameters

Forward Error Correction (FEC)

• Forward error correction (FEC) - A process by which additional data is coded with the information data that enables the receiver to detect and correct some classes of errors that occur during transport.
• FEC encodes additional data bits to the transmitted signal for error detection and correction.

Byte Interleaving

• Byte Interleaving FEC encodes additional data bits to the transmitted signal so that the receiver can detect and correct bit errors, which occur during transport.
• Advanced PHY technology includes more symbol error correction per transport blocks.
• Byte interleaving spreads short bursts of errors among Reed-Solomon (RS) FEC code words.
• Easier to repair a single bit error than a whole byte.

Adaptive Equalizers

• Adaptive equalizers were introduced in the DOCSIS 1.1 specification to compensate for linear distortions in the RF return path.
• Adaptive equalizers compensate for:
  • Micro-reflections - Reflections caused by impedance mismatches having a time difference between the main signal and the echo of less than 1 microsecond.
  • Group delay - The difference in transmission time between the highest and lowest of several frequencies through a device, circuit, or network.
  • In-channel amplitude ripple or tilt
  • Other linear distortions
• Feedback circuit samples the RF carrier and adjusts to minimize bit errors.
• The number of taps in the delay line indicates the sampling and correction capability.
• The higher the number of taps, the higher the sampling and correction.

Four-Tap Adaptive Equalizer Example

• Channel frequency response drops 4 dB at the channel's upper edge.
• Most of the channel’s energy passes through the main tap and amplifier (b0).
• A portion of the channel’s remaining energy is directed to taps that amplify only a portion of the channel’s bandwidth.
• Special algorithms adjust the gain of the taps so that, when the outputs from all the taps are summed together, the resulting channel response is flat and free of bit errors causes by linear distortion.
• After eight-tap adaptive pre-equalizers were introduced in DOCSIS 1.1, it became evident more taps were necessary to support higher modulation order schemes in the return path.
• The number of pre-equalizer taps increased to 24 in DOCSIS versions 2.0 and higher.
• Digital signal level meters (SLM) display adaptive equalizer corrections in a bar graph.
  • Each bar represents an equalizer tap within a unit of time.
  • The vertical scale represents signal reflection in decibels (dB) relative to the eighth bar (incident signal).
  • Bars to the left of the incident signal tap represent the incoming signal's frequency response and group delay.
  • Bars to the right represent decision feedback equalizers indicating the amount of equalization applied.
  • Ideally, bars gradually taper off in amplitude from the incident signal tap.
  • Higher amplitude taps compensate for impairments caused by signal reflection or distortion.

Micro-Reflection Distance Calculation

• Test instruments calculate the approximate distance to the cause of the micro-reflection based on the compensating tap.

Channel Bonding

• Channel bonding: A logical process that distributes data packets across multiple independent RF channels during transport, then combines the data packets into one higher speed data stream at reception.
• A process introduced in DOCSIS 3.0 to multiply the data throughput of a single RF carrier by the number of bonded carriers.
• A 120 Mbps data stream is distributed across four 64-QAM channels.
• Each channel has 6.4 MHz bandwidth and 30 Mbps maximum data throughput.
• The channels are transported to the headend and delivered to the cable modem termination system (CMTS).
• The CMTS combines the data packets back into a 120 Mbps data stream.

Ingress Cancellation

• Chipsets with ingress cancellation algorithms identify and remove impairments, such as ingress and common path distortion (CPD), within the upstream channel bandwidth.
• They remove CPD energy below the DOCSIS channel and intermittent interference.
• Ingress cancellation enables the use of frequency spectrum once considered unusable.
• Carrier transport in the return path is more reliable without ingress disruptions.
• Ingress cancellation chipsets are installed in most CMTS return path receivers.
• These chipsets work with all versions and models of cable modems.
• Ingress cancellation only removes ingress or CPD on a return channel.
• Ingress and CPD between return channels remain and can cause nonlinear operation of the laser and create distortions.
• Monitoring and repair of ingress and CPD remain critical for reliable return path operations.

S-CDMA Details

• DOCSIS 2.0 and 3.0 modems can be configured to operate in A-TDMA or S-CDMA mode.
• S-CDMA uses DSSS coding to randomize data before transmission and spreads it out using one of 128 code words.
• Code words and spreading algorithms are arranged in an orthogonal pattern.
• The CMTS can identify simultaneous transmission of up to 128 modems on the same RF return channel in the same TDMA time slot.
• S-CDMA coding enables the CMTS to identify modem transmission amongst random noise.
• S-CDMA was enhanced in DOCSIS 3.0 for improved performance in noisy environments at lower levels into the CMTS receiver.
• S-CDMA can be used with 64-QAM at frequencies below 20 MHz (the noisiest part of the return spectrum), with high data throughput and minimal data errors.
• Direct sequence spread spectrum (DSSS) - A spread spectrum technique that resists interference by mixing in a series of pseudo-random bits with the actual data.
• DSSS combines a data signal at the sending station with a higher data rate bit sequence, which many refer to as a chip sequence (directly related to processing gain).

Recap of MAC and PHY Technologies

• MAC ensures fair access to the return frequency band and correct data throughput.
• TDMA requires the STB to request a time slot from the STB headend controller.
• A-TDMA and S-CDMA were advanced PHY technologies added to DOCSIS 2.0.
• Ingress cancellation removes CPD and ingress within the channel bandwidth.
• The CMTS can identify simultaneous transmission of up to 128 modems using S-CDMA.

Analyzing Return Path Carriers

• Carriers on the RF return path come from STB + DOCSIS devices (cable modems, EMTAs, and cable gateways).
• Low data rate digital modulation and narrow bandwidth RF carriers were used for reliable data transport.
• Technologies introduced in DOCSIS 2.0 allow reliable reception and decoding at higher throughputs.
• Digital carriers from DOCSIS modems can have different modulation types and access technologies.
• Technicians must recognize multiple modulation and access schemes.

STB Return Carriers

• STB return carriers are sent in response to poll requests, video-on-demand (VOD) orders, and trick-mode commands.
• STB polling collects impulse pay-per-view (IPPV) data and confirms network connection.
• STB controller manages polling traffic to avoid data packet collisions.
• VOD orders and trick mode commands are customer-generated and can cause collisions.
• STB access protocols include collision recovery for retransmission of data.
• STB polling is done to collect impulse pay-per- view (IPPV) data from the STB and to periodically confirm that the STB is connected to the network.
• The Digital Video Subcommittee (DVS) of the Society of Cable Telecommunications Engineers (SCTE®) recognizes two different sets of STB upstream channel specifications, SCTE 55-1 2009 and SCTE 55-2 2008.
• SCTE 55- 1 2009 is applied to Arris (formerly Motorola) and Arris-compatible STB systems, and SCTE 55-2 2008 is applied to Technicolor (formerly Cisco Systems) and Technicolor compatible STB systems.
• Trick mode - Enables viewer to fast forward, rewind, and pause live broadcast TV, video on demand applications, and digital video recorders.

STB Upstream Channel Specifications

STB Upstream Channel Assignments

• STB upstream channels are typically between 8 MHz and 20 MHz (the noisiest part of the RF return spectrum).
• STB upstream channel specifications call for QPSK digital modulation at bandwidths ranging from 192 kHz to 2 MHz.
• Data rates are slower than DOCSIS modems.
• QPSK performs well in the presence of noise and interference.
• Lower QPSK data rates are tolerable due to less urgency for the delivery of STB data packets.

Modulation

• Quadrature phase shift keying (QPSK) - A digital modulation technique that combines two binary phase shift keying (BPSK) signals $90^\circ$ out-of-phase.

DOCSIS 1.0 + 1.1 Return Carrier Specifications

Bandwidth and Throughput

• Operators increased upstream channel bandwidth while using QPSK modulation.
• Demand for higher throughput incentivized work on the return path.
• Some operators transitioned to 16-QAM, doubling maximum throughput to 10.24 Mbps.

Modulation Technique

• Quadrature amplitude modulation (QAM) - A modulation technique for transporting digital information using a combination of amplitude modulation (AM) and phase modulation (PM).

DOCSIS 2.0 Return Carriers

• Physical (PHY) Layer - Layer 1 in the Open System Interconnection (OSI) architecture that provides services to transmit bits or groups of bits for transport between open systems entailing electrical, mechanical, and handshaking protocols.
• DOCSIS 2.0 was developed with advanced PHY layer technologies for near-symmetrical throughput.
• DOCSIS 2.0 enables higher orders of modulation and wider bandwidth return carriers.
• A DOCSIS 2.0 modem can transmit 64-QAM on a 6.4 MHz-wide carrier, for a maximum data throughput of 30.72 Mbps.

Traffic Access Schemes

• The advanced PHY Layer in DOCSIS 2.0 utilizes one of two traffic access schemes, A-TDMA + S-CDMA.
  Advanced time division multiple access (A-TDMA) - A time division multiple access (TDM) protocol with enhancements that include improved forward error correction (FEC) and ingress cancelation.
• A-TDMA is an enhanced type of TDMA, protecting data from noise and impairments.
• Synchronous code division multiple access (S-CDMA) - A coded network access technique in which the data is coded and spread across the spectrum using an algorithm that places the data in an orthogonal pattern that enables the receiver to identify and decode data that is received simultaneously.
• S-CDMA transmits data over a noisy transport path by distributing the data for longer periods of time across the bandwidth of the carrier.
• S-CDMA transmission has more average power than A-TDMA.

Maximum Transmit Levels from DOCSIS 2.0 Modems

DOCSIS 3.0 Return Carriers

• Channel bonding - A logical process that distributes data packets across multiple independent RF channels during transport, then combines the data packets into one higher speed data stream at reception.
• Channel bonding multiplies the data throughput of a single RF carrier by the number of bonded carriers.
• With four 64-QAM carriers, each with 6.4 MHz bandwidth and 30 Mbps throughput, the cumulative stream throughput is 120 Mbps ($4 \times 30 = 120$).
• Originally, two to four 6.4 MHz QAM channels could be bonded.
• DOCSIS 3.0 allows spreading upstream channels across the return spectrum to avoid unusable portions.
• DOCSIS 3.0 offers an optional expanded return spectrum plan of 5 MHz to 85 MHz.
• Cable modem manufacturers developed modems transmitting up to eight bonded channels for a maximum upstream throughput of 240 Mbps ($30 \times 8 = 240$).

Power and Bandwidth

• The amplitude of the carrier or carriers from the cable modem is adjusted and based on the cumulative input power received by the cable modem termination system (CMTS).
• When carriers of the same amplitude and bandwidth are bonded and simultaneously transmitted, the cumulative power into the CMTS increases by 3 dB for two carriers and by 6 dB for four carriers.
• Reducing the amplitude level of the transmitted carriers to maintain the cumulative power of a single carrier decreases the C/N by the same amount.