DIGITAL TRANSMISSION OF FM AND TV COMPOSITE STEREO USING DS-1 (T-1) TELEPHONE CIRCUITS
AUTHOR- CHARLES H. HAUBRICH, PRESIDENT, QEI CORPORATION
DIGITAL SAMPLING BASICS
SIGNAL SAMPLING
In the 1930's, Harry Nyquist defined a formula which states that if an analog signal is sampled at a rate greater than twice its bandwidth, all the information within it can be extracted. For example, in FM stereo, the left and right channels are band limited to15 kHz and sampled at 38 kHz. Theoretically, a 15 kHz channel could be sampled at a 30 kHz rate. However, any signal in excess of 15 kHz would "alias" and become a signal with a frequency equal to 30 kHz minus the input frequency thereby producing a frequency less than 15 kHz in addition to the original frequency. It follows then that the complexity of the input low pass "anti-aliasing" filter is determined to a great degree by how much the sample rate exceeds twice the cut off frequency. For example, BTSC (TV Stereo) is sampled at a rate equal to two times the horizontal sweep rate (31,468 kHz) which is very close to the Nyquist minimum. Accordingly, the input filters are substantially more complex than FM stereo which is sampled at 38 kHz. Since increasing the sample rate increases the bandwidth required to transmit the sampled signal, the sample rate used becomes a compromise between the complexity of the band limiting (anti-aliasing) filter and the transmission bandwidth available, in most applications.
QUANTIZATION
In order to "digitize" an analog signal, the signal level at the instant of sampling is converted to a binary number by an analog to digital converter. The number of bits in this binary number determines the number of levels that the signal will be "quantized" into. Because the information that lies between these levels is lost, the number of levels used determines the dynamic range and distortion of the recovered analog signal. If eight bits are used, 256 levels can be determined. This yields a dynamic range of about 48 kB and about 0.5% distortion at full output. Each bit added doubles the number of levels, increases the dynamic range by 6 dB, and decreases the distortion by a factor of two. For example, 14 bits yield 16384 levels, a dynamic range of about 84 dB, and about .01% distortion.
BANDWIDTH
The bit rate of an uncompressed digital signal is equal to the sample rate times the number of bits per sample. For example, in order to transmit an analog channel with a bandwidth of 10 kHz, a sample rate of 24 kHz, and a dynamic range of 84 dB (14 bit quantization), a bit rate of 336000 bits per second is required (24000 X 14 = 336000). If the same channel required only 48 dB dynamic range and 0.5% distortion, only 8 bit quantization would be needed. This would reduce the bit rate to 192000 bits per second.
SAMPLE RATE MULTIPLICATION
When a digital bit stream is fed to a Digital to Analog Converter, the resulting analog signal does not vary smoothly. Rather, it jumps to the appropriate level at each sample (See Fig. 1). The spectrum of this signal consists of the original analog signal and upper and lower sidebands about the sampling frequency and its harmonics (See Fig. 2). An analog low pass filter similar to the input "anti-aliasing" filter is used to eliminate the spectrum above the highest frequency to be passed by the channel (See Fig.3 & 4). This frequency cannot be greater than one half of the sampling frequency or else aliasing will occur.
In order to transmit a digital signal using the minimum bandwidth, the sampling rate must be held to a minimum. This requires the use of a sharp cutoff analog filter after the Digital to Analog Converter. It is very difficult if not impossible to manufacture a stable sharp cutoff analog filter without phase (time) non-linearities. This means that all program frequency components do not take the same amount of time to get through the filter. The resulting signal contains artifacts which may degrade the sound of the program. If the sample rate could be multiplied after the digital signal was received, thereby using minimum transmission bandwidth, the output analog filter could be greatly simplified. Fortunately, Digital Signal Processors (DSP) allow a technique using Finite Impulse Response (FIR) filters to increase the sampling rate before the bit stream is fed to the Digital to Analog Converter. An FIR filter function is performed by computation in the digital domain. It is possible for an FIR filter to have an extremely sharp frequency cutoff characteristic with zero group delay (linear phase). The DSP processor looks at two adjacent samples and computes the value of the sample that would be between them, based on the frequency characteristic of the FIR filter function. Therefore the bandwidth of the original analog channel is maintained but the number of samples (sample frequency) is doubled. This technique can be repeated up to the speed capability of the components involved.
TELEPHONE NETWORK DIGITAL TRANSMISSION
HISTORY
T1 was implemented by the Bell System in the early 1960's as a means of increasing the capacity of existing trunks between Central Offices. Research performed at Bell Labs led to the technology which allowed 24 voice channels to be digitally transmitted in one direction using Pulse Code Modulation (PCM) on a single twisted pair. The required electronics were much cheaper than installing more cable and since the PCM transmission was much less susceptible to noise and crosstalk, audio quality improved dramatically as well. Since then, the system has evolved to where thousands of voice channels are transmitted simultaneously by PCM using light on a fiber the size of a hair.
SYSTEM SYNOPSIS
A 3 kHz analog voice channel is first band limited to around 3300 kHz. The signal is then sampled at an 8 kHz rate. Each sample is quantized to 8 bits. This means that a voice channel requires 64000 bits per second to transmit. This 64 Kbit signal is called a DSO channel. 24 DSI channels are combined along with 8000 bits per second for framing to make a DS1 (T1) signal. The equipment that accomplishes this is called a Channel Bank. The DS1 (T1) signal has a bit rate of 1544000 bits per second (1.554 Mbps) [(64000 X 24) + 8000] . It was found that a telephone cable twisted pair could accept this high a bit rate provided the terminating impedance was lowered to 100 ohms and regenerative repeaters were installed about every 6000 feet. These repeaters decide whether the degraded incoming signal is a one or a zero and send a perfect one or zero to the next repeater. In this manner, the bit stream at the far end becomes exactly like the input bit stream. Since the analog signal is encoded in the numbers represented by the digital bit stream, the distortion and noise of the recovered analog signal is not affected by the transmission medium so long as there are no bit errors. DS1 signals can be multiplexed with other DS1's into higher rate signals for transmission via other media, e.g., coaxial cable, microwave, and fiber optic cable. DS1 is the lowest rate that can be multiplexed asynchronously. As long as the signal is within specifications, the telephone network multiplexers will lock on the framing pulses, transport the signal through the network and hand it back at the same frequency it was received. In order to accomplish this, the subscriber's equipment must generate the signal in accordance with stringent requirements. The frequency of the bit stream must be 1544000 + or - 75 bits per second. The pulse shape must conform to FCC rules. A zero is represented by zero volts; a one is represented by either plus 3 volts or minus 3 volts (opposite polarity of the preceding one). This is called bipolar signaling or Alternate Mark Inversion (AMI). This type of signaling places a null in the signal at DC allowing power to be sent on the line. It also allows timing to be extracted from the bit stream provided there are no long strings of zeros. 12.5% of all bits must be ones, and there can be no more than 15 zeros in a row. These are known as "ones density" requirements and are necessary to insure that the network repeaters do not jitter and lose sync. Framing pulses must conform to a pattern known as D4. This pattern is recognized by the network multiplexers and is required if the DS1 signal is to be multiplexed with other DS1 signals for transmission over microwave or fiber optic cable.
BROADCAST CONSIDERATIONS
Even if you are currently leasing 8 kHz or 15 kHz analog pairs, the telephone company probably transports them digitally between central offices. Some broadcast engineers have heard the term T1 in relation to this and in some cases complain about the sound of the programming. Most likely, this is caused by the digitizing process used by the telephone company. In order to fit the analog signal into a discrete number of DSO channels, the sampling rates are multiples of 8 kHz. Thus, a 15 kHz circuit is sampled at 32 kHz. This requires an extremely sharp filter with attendant phase distortion. Amplitude and phase tracking between stereo pairs can also become difficult to maintain.
Since the early 1980's, DS1 service has been available to individual subscribers as a tarriffed point to point dedicated service. Virtually any twisted pair can be configured as a T1. It is a four wire service, i.e., bi-directional, one pair each direction. It can be ordered as a point to point DS1 or T1 circuit. DS1 and T1 tend to be used interchangeably. It may also be called 1.5 Megabit service. AT&T's trademark is "ACCUNET 1.5". Southern Bell's trademark is "MEGALINK" just to name a few.
DS1 circuit prices vary widely. At the present time, a DS1 costs about the same as a 15 kHz stereo pair plus a control circuit in many places. However, factors other than increased audio quality weigh on the side of the DS1. First, it allows additional AM or SCA quality audio channels to be transported along with the Composite Stereo channel without crosstalk. Secondly, DS1 circuits are bi-directional. This capability allows RPU or Satellite antennas and receivers to be installed at the transmitter site. The RPU or Satellite audio can then be transported to the studio without degradation for production or mixing. The costs of analog lines necessary for these additional functions make the DS1 very cost effective. The trend is to decreasing costs for digital service and greatly increasing costs for analog service. In addition, since the telephone networks are digital, testing and repair of DS1 circuits is generally much quicker than analog.