Rigid radio

We are observing a massive increase in wireless connections in daily life, especially in the area of home automation where more and more remote controlled garage doors, heat cost allocators, window blinds and other devices are now employing wireless technology for the sake of convenience and to save valuable energy.

With an increasing number of wireless connections comes the increased risk of data packet losses, because every other channel in the same frequency band is also an interferer for the actual transmission. This is not a major concern if we implement corrective measures, such as checking whether the package actually arrived at the receiver and re-attempt transmission if necessary. Other possibilities include encryption techniques such as multi-carrier spread spectrum method or DSSS (Direct Sequence Spread Spectrum), which is preferably used in cellular networks such as UMTS.

Figure 1: Comparison of two transceivers without interferer.

All these methods do not alter the fact that the input level of the receiver increases with a greater number of neighbouring stations. These neighbouring stations do not necessarily work at exactly the same frequency; a channel interference even a few 100kHz away from the wanted signal carrier can affect the capabilities of the receiver. In addition to the radiated power of the interferer the frequency spacing between receiver and interferer influences the error rate in the transmission. Figure 1 shows this behaviour without any interferer. Adding a transmitter in a neighbouring channel will interfere with the wanted signal which is shown in Figure 2.

Figure 2: Impact of an interfering radio transmitter (in red) on the wanted signal.

A transceiver with a blocking performance 30dB better than another device can achieve a range improvement by a factor of 32. This can be shown — and is well-known — that an improvement of the link-budget by 6dB is equivalent to doubling the range. Or, vice versa, a potential interfering signal could be closer to the receiver by a factor of 32.

Adjacent channel attenuation

The transceivers from Texas Instruments’ ‘Performance Line’ show very good performance with respect to adjacent channel attenuation and input sensitivity, when compared to other transceivers. Other manufacturers can produce transceivers with similar values but a direct comparison of the IC parameter is often not easy because the values ​​in the data sheet are often obtained with different measurement conditions, such as data rate, modulation type, frequency deviation, or filter settings.

For a better understanding, the following example shows the explanation of the term sensitivity and blocking, which is the same as adjacent channel attenuation. The data sheets of two different 2.4GHz transceivers specify, for example, a sensitivity of −100dBm vs −97dBm, and blocking of 27dB vs 55dB at ±5MHz offset.

Considering only the sensitivity, the first device would be the better choice; a wireless LAN signal at a frequency distance of 5MHz away with a power level of -80dBm will not disturb either receiver. Both are below the sensitivity of the receiver and represent no impairment on performance.

If the wireless transmission power increases to -40dBm (for Germany the maximum radiated RF power in the 2.4GHz band is specified at 20dBm), this calculation looks quite different; in this case the second device is the better choice because it offers a 28dB better sensitivity (with the strong interferer).

Figure 3: Identical measurement conditions for both transceivers.

To prove this, comparative measurements were made with the CC1120 and CC1101, two transceivers operating in the 868MHz ISM band. The test should prove that the performance transceiver CC1120 is less sensitive to a jammer compared to the CC1101 transceiver. Two SmartRF05EB evaluation boards were fitted with 2 CC1120EMs and then with 2 CC1101EMs. An un-modulated carrier frequency with different frequency separations to the desired signal was used as a jammer. The transmitter board was operated without an antenna, so that the received power could be limited to approximately -90dBm. All boards have been configured with the SmartRF Studio software (www.ti.com/SmartRF). The power level at the receiver, as well as the measured Packet Error Rates (PER) were measured with this tool. Configuration parameters for the transmitter (TX) and the receiver (RX) were: carrier frequency = 868.0 MHz; data rate = 38.4kbit/s; modulation = 2-GFSK

First, the test was performed with the jammer turned off. The evaluation boards were placed so that the receiver measured a signal strength of about -90dBm without any transmission errors. In all tests the Packet Error Rate was determined on 1000 transmitted data packets. Using the Performance Line transceivers an increase in the Packet Error Rate was observed when moving the source of interference, to around a meter distance. The Packet Error Rate increased continuously when the distance to the interference source with the CC1101 was within 8m. The results are summarised in the Table 1.

Table 1: Summary of the Packet Error Rates observed with distance.

In summary the measurements prove that the performance line transceivers are more robust against interfering frequencies and jammers. A jammer at a frequency distance of 200kHz could be placed about four times closer to the performance line transceiver compared to the conventional CC1101.

Author profile: After studying communications engineering in Cologne, Hans-Guenter Kremser worked for several years as a development engineer and in various semiconductor manufacturers as an FAE. Today he works as a Principal-FAE (Field Application Engineer) for analog products at Texas Instruments.