|
Two-Way RF Technology Targets Low-Power Wireless Apps
By Jeff Robillard and Jim Wight
New protocol technique enables the development of two-way short-range
wireless systems that offer improved performance, better reliability, and longer battery life.
Every year, wireless technology finds a place in new applications. From mobile environments to the home, wireless technology has become a bigger part of an end user’s life.
There is no question that many design dollars are being spent on improving the performance of high data-rate wireless systems, such as commercial cellular and PCS systems. But there is also a clear call to improve the RF
technology used to develop low-power, low bit-rate wireless systems, such as garage door openers and security systems.
As wireless technology becomes more entrenched in low-power, low-bit-rate systems, end-users will increasingly demand higher quality and reliable RF links. These systems tend to be control-oriented and are extensively employed in the automotive, home and building automation, and personal computing sectors. As the growth of these applications introduces more traffic into the airwaves,
designers must develop systems that maintain reliability even in congested frequency bands.
Today, the lion’s share of short-range wireless systems provide one-way, unidirectional RF or IR transmission and aren’t equipped to handle the congested frequencies of tomorrow. That trend is changing. Two-way, bidirectional (duplex) systems are being developed for low bit-rate wireless applications. These systems ensure reliability while providing a foundation for the implementation of new features in
next-generation products.
Protocols are key
Prior to the advent of today’s packet-based networks, point-to-multipoint RF satellite communication systems distributed data over a single radio channel or frequency. This allowed multiple earth stations to communicate to each satellite over a single radio channel. These two-way RF systems were the first to employ packet-like
protocols to achieve reliable transmissions.
Today’s low-power, low-cost RF systems can employ similar techniques to enable highly reliable links. RF links are often subjected to interference from adjacent frequencies or channels that disrupt transmissions. The use of a two-way system, in combination with simple lightweight protocols, minimizes disruptions and increases reliability. Two-way transmissions allow an acknowledgement (ACK) function, system self-optimization, authentication algorithms for
high security, error detection and correction, and return of status and alarms.
Consider a packet-like implementation that consists of application software, a protocol stack, and physical layers. To minimize complexity and cost, application and protocol stack layers should be lightweight.
Figure 1
illustrates how a lightweight protocol is implemented with an RF transceiver, which serves as the RF portion of the physical layer. The RF transceiver should
maximize flexibility when selecting an appropriate protocol for a specific application, or when back-fitting onto existing protocols. The RF transceiver includes the transmitter, receiver, channel switching, and selectivity/sensitivity optimization.
A microcontroller, ASIC, or field-programmable gate array (FPGA) is employed to implement the baseband portions of the physical layer and the upper layers of the protocol stack. The controller, ASIC, or FPGA could also include algorithms such as carrier sense
and collision detection, link maintenance, and error correction. The selected protocol should be lightweight to minimize the complexity and cost of the microcontroller, ASIC, or FPGA.
Effective packet-based protocols employ two-way transmission where the originating end always expects an ACK from the destination end. The originating end retransmits packets until an ACK packet is received. This is not possible with today’s one-way protocols, which are essentially open-loop systems. Two-way
transmission provides a closed-loop operation that detects transmission problems and permits self-optimization for improving reliability at the system level.
Configurability means reliability
The key to developing efficient two-way RF systems lies in the protocol stack. By employing an efficient protocol, engineers can improve the overall performance of two-way RF systems.
Engineers can choose from a plethora of protocols when developing two-way RF systems. One of the newest and most efficient, however, is a ping-pong-ping protocol.
Ping-pong-ping is a packet-like protocol that takes advantage of two-way transmission to maintain links and ensure reliable operation. With an established link, the originating end guides the destination end through an optimization process. By issuing commands, the originating end measures the link’s performance and controls
physical RF resources at both ends to improve and optimize features such as channel or operating frequency, transmit power, sensitivity, and selectivity.
Figure 2
illustrates an example of an automated maintenance routine generated by a ping-pong-ping protocol scheme. The originating and destination ends are referred to as master and slave, respectively. The algorithm selects a channel with the highest level of performance among a fixed range of available
channels. Both ends implement a lightweight algorithm for measuring the link’s performance.
In
Figure 2
, the master sends a command to test the uplink, and readies itself to receive a prescribed transmission containing the necessary patterns to derive a performance measurement. The slave receives the command and responds by sending a prescribed transmission. The master then receives the transmission and measures the performance of the uplink. Once this is
complete, the master sends a command to the slave to test the downlink. The slave receives the command, readies itself for the prescribed transmission, and sends an ACK.
The master receives the ACK function and sends the prescribed transmission. The master now waits for a return transmission of the downlink’s performance data. The slave receives the transmission and measures the performance of the downlink. After completing this task, the slave sends the measured performance data of the downlink.
The master receives the performance data of the downlink, determines if the link is satisfactory, and then decides on the next action. If the master is satisfied with the performance, no further commands are issued. If the master is dissatisfied, it issues another command to switch to another channel. The process then repeats itself until satisfactory performance or best-effort performance is achieved. Similar processes could be employed for adjusting transmit power, sensitivity, and selectivity to
gain optimal link performance.
A practical example
The one-way systems that are predominantly in use today are prone to reliability problems that inconvenience the user. Consider a garage door opener where the RF environment is noisy at the time of use.
In a one-way system, the user presses a button on the remote. If the garage door doesn’t open, the user presses it
repeatedly until it opens. Additionally, in a rolling-code-based system, both ends fall out of synchronization on the rolled codes and cause a lockout. In some cases, it is an elaborate process to get both ends resynchronized — not very convenient for the user.
In a two-way system employ-ing the ping-pong-ping protocol scheme, when the user presses the button on his or her remote, the remote immediately prepares for an ACK response from the garage door opener control unit located in the
garage. If there is no response, the remote retransmits. This sequence repeats itself until the garage door opens. Since the sequence is automatic and transparent to the user, the perceived reliability of the system increases.
The non-coherent approach
There is a long-standing concept that frequency-shift keying (FSK) links provide superior reliability performance. Many wireless
systems employ complicated coherent (synchronized) transmitters and receivers that maximize the throughput of data in a limited channel bandwidth or in situations requiring multiple access. However, many wireless applications do not require maximum data throughput or simultaneous multi-user traffic. In these systems, the use of coherent transmitters and receivers translates into an increase in cost and power, without necessarily providing any improvement in performance over simple non-coherent transmitters and
receivers.
A coherent communication link, such as FSK, requires a transmitter with a precise carrier frequency, and a receiver with circuitry able to reproduce this frequency exactly. Both of these functions require the use of phase-locked loops (PLLs) to synthesize the transmitter carrier and to synchronize the receiver reference. In terms of size-on-chip (cost), power, and sensitivity to noise and interference, the loops used to synthesize the carrier and synchronize the reference dominate all
other circuit functions in the transmitter and receiver.
Non-coherent communication links, on the other hand, employ modulation schemes that do not require precise frequency control in the transmitter, or precise reference generation in the receiver. As such, they perform well in channels experiencing fading, blocking, or doppler. One modulation scheme that can be used in non-coherent systems is on-off keying (OOK).
OOK provides robust detection in the presence of fading interference or
doppler, since there are no sensitive loops to be disturbed. Furthermore, where the channel bandwidth exceeds the signal bandwidth, a strong improvement in receiver signal-to-noise ratio (SNR) can be achieved by sampling the received “on” pulse several times over its duration and adding the values. This pulse integration technique will improve the SNR by a value between
n
and
 , where
n
is the number of samples. Other refinements to the pulse
integration technique include the “figure skating” secondary detection criteria, in which the largest and smallest values of the samples are discounted.
OOK also allows for the use of imaginative techniques such as the RF Manchester encoding scheme. Because this scheme encodes each 1 as an on-off symbol, and each 0 as an off-on symbol, post-demodulation timing is not compromised by a long string of 1s or 0s. The use of such split-bit symbols allows the receiver to ensure that at least one
transition occurs each bit interval in order to establish timing. Other novel encoding techniques can be envisioned with the use of OOK signaling, such as pulse-width modulation (PWM), where the duration of the “on” portion within a symbol period can be varied to convey a 1 or a 0.
Longer battery life
In handheld appliances such as remote keyless entry (RKE) products,
reliability is often associated with battery life. If a remote does not work because of a dead battery, it is assumed to be unreliable, especially if the batteries need to be changed often. Battery life is proportional to the number of transmissions, transmission length, and how many times the remote’s buttons have been pushed.
As discussed earlier, low bit-rate, one-way wireless systems often suffer from unreliable transmissions because many devices are simultaneously contending for the same
operating frequency, or are bombarded with interference from transmissions at adjacent frequencies. With one-way systems, it can be particularly difficult to implement an effective protocol and achieve reliable transmissions under such conditions.
Consider the case of RKE systems. Various techniques are employed to improve reliability in these systems, such as multiple transmissions of the same data packet, and forward error correction. The length of transmissions is extended because of the
additional redundant information that must be transmitted, which can consume valuable battery power in portable applications. Even with the employment of such techniques, there is no guarantee that the transmission will succeed. In frustration, users will repeatedly push buttons on their remotes in an attempt to get the transmission through. Such activity ends up consuming even more battery power. The remote’s batteries wear out sooner, thereby inconveniencing the user.
By employing a ping-pong-ping
technique, the number of transmissions, the amount of redundancy, the transmission length, and the number of times buttons are used on the remote can be reduced. This extends battery life. RF vs. IR
Two-way IR systems, such as beaming between PDAs, provide unreliable transmissions. Generally, these links are highly directional, require line-of-sight operation, and are sensitive to distance. Unless the PDAs are within a very short distance of each other and are well aimed at each other,
transmissions are unreliable and likely to fail.
Two-way RF transmission makes reliable beaming a reality while minimizing cost and power consumption. Consider multiple PDA stations that are RF-based — RF provides a superior range and travels through and around objects. A ping-pong-ping style protocol can be employed that allows point-to-multipoint communication resulting in the sharing of information between many PDAs. Multiple PDA stations may employ the same or adjacent radio channels as shown in
Figure 3
.
In
Figure 3
, one PDA station communicates with another by waiting until the radio channel is idle. If idle, the station sends a packet of data that includes source and destination addresses. All idle stations monitor the radio channel and only accept packets that are addressed to them.
If the station is not idle, the destination or receiving station sends an ACK signal indicating that it successfully
received the packet. If the source or transmitting station does not receive the ACK signal within a prescribed period of time, it assumes the transmission has been corrupted and retransmits the packet.
Stepping ahead
One-way systems often suffer from unreliable transmissions due to interference or frequency contention, which can lead to ineffective operation and short battery
life in handheld applications. With the use of appropriate protocols, two-way systems can help remedy these problems, thereby increasing the utility of these applications. Furthermore, by employing a simple layered protocol, it is possible to build point-to-multipoint applications that allow multiple devices to share information.
Though the examples presented here are simplistic, they do illustrate the benefits of a two-way RF system. Similar RF systems can be implemented with simple, lightweight
protocols. Applications that are good candidates for lightweight protocols include high-security residential keyless entry products, garage door openers, security systems, home and building automation control systems, wireless message boards and displays, automatic meter reading devices, telemetry systems, wireless PC peripherals, game consoles, PDAs, medical systems, industrial systems, automatic identification products, and data collection systems.
Jim Wight
is a
senior radio architect at Philsar Semiconductor. He is also a professor and head of the electronics department in the School of Engineering at Carleton University, Ottawa, Canada. He holds a BSEE from the University of Calgary as well as an MSEE and PhD in electronics engineering from Carleton University. Wight can be reached at
jimw@philsar.com
.
Jeff Robillard
is a product manager for Philsar. Prior to joining Philsar, he held various product
management and sales positions in the telecommunications business, including positions at Mitel Semiconductor and Texas Instruments. Robillard holds a BSEE from the University of Waterloo in Waterloo, Ontario, Canada. He can be reached at
jeff_robillard@philsar.com.
.
| 
Puzzled by a network processing design issue?