university of luton   CABLE MODEMS AND OTHER BROADBAND TECHNOLOGIES
     

NAVIGATION

Home of Cable Modems

My Investigation on Cable Modems

Investigation of CM's operation & architecture

Cable Modem Standards

Alternative Broadband Technologies

The Future of Cable Modems

Photos of Cable Modems

Other CM Links

 

2.           Investigation of technical operation & Architecture

   of Cable modems

2.1       Cable Modem’s Architecture

2.1.1    Tuner

2.1.2    Demodulator

2.1.3    Burst Modulator:

2.1.4    MAC (Media Access Control) device

2.1.5    Microprocessor

2.2       Technical Operation of Cable Modem Systems

2.2.1               Initial modem setup

2.2.2    Setting Reverse Path

2.2.3    Setting Transmit Levels

2.2.4    Timing Offset (Delay)

2.2.5    IP address assignment

2.2.6    Setting Carrier levels

2.2.7    Other Issues

 

 

CABLE MODEM’S ARCHITECTURE

 

Cable modems brands may vary but all have the same basic architecture. The major components are five. They and their interconnections can be seen below and explained further:

 

Pic.2 Block diagram of  the organization of the key elements inside a typical cable modem.

  

2.1.1 Tuner

 Tuner is the device that connects the cable modem to the cable outlet, usually with the addition of a splitter. The splitter is used to separate the internet data channel from normal CATV programming.

 The operation of the tuner is simple. It receives the modulated digital signal through, an otherwise unused, cable channel and passes it to the demodulator. In most of the cases the tuner contains a diplexer which is used to make use of two different set of frequencies, one for the downstream and the other for upstream. The downstream set of frequencies is between 42 -850 MHz while the upstream is between 5 – 42 MHz. On the other hand modems that don’t have diplexer use the tuner for downstream data and a dial-up modem for upstream traffic.

 Furthermore a new concept of a silicon tuner has been developed over the last years. It is basically a tuner on chip. The big advantage of this new technology is the reduced cost.

 

 2.1.2 Demodulator

 The demodulator operates by converting the radio-frequency (RF) signal and its wave variations into a signal that can be accepted by the A/D converter. Then the A/D take the analogue signal which varies in voltage and turns it into a digital signal where the only appropriate values are 1’s and 0’s.

 

Moreover an error correction device is used. It checks for errors by comparing the received information against a known standard. After conversion and error correction is finished, the data consist of network frames. These frames might be in compressed format, in most cases MPEG format. Compression helps the cable system to send, up to 10 times more data within a single 6 MHz channel.

 

2.1.3 Burst Modulator:

 The reverse of the demodulation process is the modulation.  The burst modulator is used to convert digital network data back to radiofrequency signals. In the transmit direction it sends the data to the tuner. Its operation is based on Reed – Solomon encoding and a D/A converter. First the burst modulator does Reed-Solomon encoding of each burst, then modulates the QPSK/QAM – 16 on the selected frequency and finally it does the digital to analogue conversion. The output signal is then feed through a driver with variable output level.

As Integration “race” drives more and more functions into a single chip, a combined demodulator and burst modulator has been developed and is available in the market.

  

2.1.4 MAC (Media Access Control) device 

 The MAC acts as an interface between upstream and downstream traffic in a cable modem. This device handles hardware and software network protocol communication. All network adapters contain a MAC layer but in this case cable modem’s MAC is pretty complex compared to Ethernet’s MAC.  For this reason help from the modems microprocessor is needed in order to handle all conversions. The microprocessor contained in the cable modem is designed to cope with this and support these MAC functions when needed without affecting performance.

  

2.1.5 Microprocessor

 The microprocessor’s job largely depends on whether the cable is designed to provide internet access or to be part of a larger computer system. If it is used for internet access with no additional computer support, the microprocessor picks up MAC's slack. On the other hand in situations that an attached computer in needed the microprocessor provides support to the dedicated MAC module. In both cases the Motorola PowerPC processor is one of the best and most common choices for designers.

  

2.2 Technical Operation of Cable Modem Systems

 In this point of the actual working and transmission operation of cable modems systems will be explained. To properly understand its operation we will follow step by step all the processes that take place in such system

 

2.2.1 Initial modem setup

 First of all when a cable modem is powered up it needs to be setup in a way that does not disturb other cable modems currently operating in the network. There are various ways to achieve this but cable modems have adopted the handshaking technique. When a cable modem is powered up it scans the Cable TV system forward path for the QAM carrier. After finding the carrier it detects its level of modulation. The system could be utilising QAM 16, 64 or 256 modulation.

 Furthermore higher orders of QAM are wanted since they offer higher speeds over the same bandwidth. Moreover after the system’s QAM is identified, the cable modem has to synchronise itself with this carrier, this operation is sometimes called QAM lock. Once this lock is established, the Cable Modem continues its handshake operation by setting up itself in the Reverse Path.  

Pic.3 is a flowchart of a typical cable modem initialisation

 

2.2.2 Setting Reverse Path

 

As we saw above the cable modem needs to initialise transmission with the CMTS at the Head End. To achieve this, the modem has to establish the “Reverse path frequency”. The Upstream Channel Descriptor (UCD) data pack contains all the needed information to establish the reverse path frequency

This data packet provides all the information for the cable modems about where to set their Reverse path transmit frequency.

  

2.2.3 Setting Transmit Levels

 Transmit Levels are very important for a cable modems system since a too high signal level will cause the cable modem to distort. Also it will overload the return path amplifiers which might cause distortion of the data. The modem needs to send its signal to the CMTS. To avoid these problems the modem begins to transmit on the Reverse Path frequency at its lowest transmit power level, gradually increasing its power until the modem can be recognised by the CMTS at the Head-End.

 At this point a two-way communication is established. The transmit level is at an optimum level and stays as it is. A Cable modem is specified to transmit in the Reverse Path at a level of 92 dBU to 112 dBU but sometimes cable modems transmit in higher levels due to improperly balanced Reverse Path and cause distortion of the transmitted Data packets.

  

 2.2.4 Timing Offset (Delay)

 Next step after forward and reverse path transmissions are established, its time to introduce the timing offset to the modems operation. The timing offset is some kind of delay needed between transmissions of two cable modems. This delay is needed to ensure that the data packets transmitted between the two modems do not collide.

  

 2.2.5 IP address assignment

 In a network each computer is assigned a unique IP address. This address is very important in a network since it is used to help the CMTS identify each computer in its network. As a result from this data can be specifically sent to the particular cable modem that requested them. More specifically data packets coexist in the system but each cable modem extracts only the ones that are “marked” with its IP address, otherwise the CMTS would have to broadcast the data to all modems on the network reducing the network’s performance by far.

 The IP address is assigned to a cable modem by sending a DHCP request as well as some other important information to the CMTS. After these procedures are finished the last step for the modem is to request the current time and date, required for time-stamping logged events, from the Time-of-Day (ToD) server.

Pic. 4  The transparent IP traffic through a cable system. 

 

2.2.6 Setting Carrier levels 

The cable modem uses an asynchronous transfer method. As a result it transmits its data as digital signal bursts. Two important factors in this are the video and carrier carriers. The video carrier is more than 14 dB higher than the audio carrier and at least 20dB more than any other signal component.  Furthermore this makes easy the setting of the carrier levels under the condition that you know the level of the video carrier. 

The digital signal maintains a fixed high level of signal power. As a result the recommended digital carrier’s average power level should be maintained at 6 dB lower than Analogue Carrier Levels. 

 

2.2.7 Other Issues

 It is also important to note that sometimes the transmitted signal from the cable modem can be so strong that any TV sets connected on the same string might be disturbed. The isolation of the splitter may not be sufficient, so an extra high-pass filter can be needed in the string that goes to the TV-sets.

  The high-pass filter allows only the TV-channel frequencies to pass, and blocks the upstream frequency band. The other reason for the filter is to block access in the low upstream frequency range from the in-house wiring.

 

< Cable Modem Investigation - Cable Modem Standards >