Antenna Design

Staying profitable is no longer a burden that falls solely on the shoulders of upper-level management. Everyone, including design engineers, must carry the load, especially when it relates to quality of service. Subscribers have high expectations, and performance issues such as blocked and dropped calls as well as poor audio quality will not go unnoticed.

Usually, these problems are related to some form of interfering signal affecting the RF link. A call can fail in any analog or digital system when the interference reduces system capacity or if the system's handoff performance is adversely affected by an interfering signal. Interference also can reduce a call's audio quality in analog and TDMA systems or block the call if at a sufficient power level. For CDMA systems, co-channel and adjacent channel interference can increase the noise floor in the receiver, which leads to fewer potential calls, higher subscriber transmit power and shorter battery life. Any interference source (co-channel or adjacent channel) can lead to frustrated customers, higher churn rates and lower revenues.

One common interference problem is an unwanted signal transmitted or received through one of the upper sidelobes of a base-station antenna's vertical pattern. But these interference issues do not have to limit your profits any longer. There are several antenna products available with patterns that feature upper-sidelobe suppression (USS). As service demand grows, networks are reaching capacity, which means that more systematic interference problems are reported. Using an antenna with USS will improve link performance and help you meet customer demand for service quality.

Several suppliers offer USS products for most frequency bands, modulation schemes, air interfaces and diversity methods. To ensure that you select the proper solution for your interference problems, you need to understand the cause-and-effect relationship between the pattern shape and other RF parameters of an antenna with USS. You also will need to understand proper antenna design principles and selection criteria.

Radiation Pattern Basics A typical directional base-station antenna (an array antenna) is composed of a group of similar elementary radiating elements. The elements can be dipoles, slots, patches or any other basic radiator. Most base-station antennas are single-column linear arrays. Both the intended signal and the interfering signals are either radiated or received through the base-station antenna's radiation pattern. An antenna's pattern is an engineer's picture of how intense the radiated energy will be in particular angular directions. Think of the pattern of a single element as the ripples in a pond after a pebble is thrown in the water. The pattern of an array is a superposition of the radiation of all the elementary radiators that compose the array. The position, intensity and relative electrical delay imposed by the feed network of each element create the shape of the pattern. The array pattern resembles the ripples in a pond when several pebbles are thrown in the water.

Consider the radiation pattern as a graph of the energy's intensity vs. angle along a plane through an axis of the antenna. A measure of this intensity is called directivity. It is a measure of the angular focusing ability of the antenna. Gain, on the other hand, takes into account all of the practical RF losses associated with an antenna. While directivity increases as an antenna's aperture area is increased, gain generally only will increase up to a point and then actually can decrease with increasing area. The relationship of gain to area is dependent on the feed network technology you implement.

The pattern of a perfect base-station antenna would have only a main beam and no sidelobes. A practical antenna has a main beam where most of the energy is focused and a series of sidelobes surrounding the main beam. But realistically, there always is some amount of energy radiated in sidelobes. The upper sidelobes refer to angles generally above the horizon of the antenna at angles greater than the main beam. Fortunately, advanced array techniques allow you to design base-station antennas with many useful features that improve the coverage and interference aspects of a network. This is possible with antenna synthesis techniques. Consider the pond analogy: The pebble thrower can tailor a ripple pattern in the pond by adjusting the position, time and speed of the pebble toss into the water. Similarly, you can achieve a desired pattern shape by adjusting the position, RF intensity and delay of each element of an antenna array.

Deployment Issues In addition to understanding how to design the pattern of a base-station antenna, engineers need to investigate the requirements for USS. There are numerous base-station siting situations that cause interference to enter an upper sidelobe. For example, if sites are located on hilly terrain, then one site may be at a higher elevation than another. This occurs with gradual sloping landscape as well. Some networks require both high-tier and low-tier antenna sites, which can create problems. In other cases, the upper sidelobes may be pointed at a neighboring building that already has its own base-station network inside. The local zoning process can create other difficult siting situations.

Also, in high-capacity urban environments, the cell radius is smaller, which means you have to either lower the height of the antennas or point the beam downward from higher positions to cover city streets. The beam can be tilted either mechanically with brackets, electrically by tailoring the aperture distribution or ideally with a combination of the two. Downtilting an antenna's main beam will lower the angle of the upper sidelobes, which can create interference problems. This can exist with or without downtilt -- for example, when the base station is in a valley and the subscribers are travelling along mountain roads in neighboring cells surrounding the base.

Antennas with USS All of these interference sources can steal caller capacity from a network. One way to combat this problem is to deploy antennas with reduced upper-sidelobe levels. Usually, the sidelobe above the horizon and nearest the main beam is referred to as the first upper sidelobe. The next one away from the main beam is called the second upper sidelobe and has lower intensity. The third sidelobe is even lower in intensity and so on.

Usually, the first upper sidelobe creates the most interference problems because it has the highest intensity and points in the most problematic directions. In some cases, you will need to reduce only the first upper sidelobe. However, there are cases where an interfering signal will enter a second or third upper sidelobe. Therefore, you may need to deploy an antenna that will lower interference in all of the upper sidelobes. There are antenna products available for both of these situations.

The definition of a sidelobe level is the relative intensity level of the pattern between the peak of the main beam and the peak of the sidelobe in question. Theoretically, an antenna without USS will provide a 13dB first upper-sidelobe level. Practical antennas without USS can exhibit an 11dB or 12dB sidelobe level. There are siting cases where only 15dB upper-sidelobe levels will mitigate interference effects sufficiently. In other cases, several more dB are required to provide acceptable performance. And in a few cases, operators have requested as much as 25dB sidelobe levels to handle extremely difficult sites. Figure 5 on page 44 illustrates an example pattern of an antenna with USS. All antennas with USS should exhibit these pattern characteristics at all frequencies across the operating band.

To help understand how a sidelobe-level specification can arise, consider the link budget and coverage issues of a typical siting configuration. Use network design principles as described in Mobile Cellular Telecommunications (W.C.Y. Lee, McGraw Hill, 1995).

In most systems, the base-station transmitter power level received at its own cell boundary is set at -100dBm. This leads to a ratio of carrier power-to-noise power (C/N) of 18dB. These numbers come from drive tests that lead to acceptable call quality at the cell boundary.

For this example, assume a cell reuse factor of seven. The cell reuse factor describes the repetition pattern of frequency use from cell to cell implemented, which maximizes the available frequency spectrum while minimizing co-channel interference. The co-channel interference reduction factor is defined to be q = u3N = u21 = 4.6 where N is the frequency reuse number of seven. Assume a 40dB/decade path-loss attenuation model for the surrounding terrain.

Now compute the interference-signal-level reduction at the co-channel cell to be 40 x log(q) = 26dB. You can obtain further signal-level reduction by downtilting the main beam. But consider site down-tilting configurations carefully because path loss often is a function of the terrain. If path loss to the cell boundary is 40dB/decade but is 20dB/decade to the co-channel cell (as might be the case over a body of water), there is an increase of 13dB in the co-channel interference signal ( 20 x log(4.6) ).

By reducing the upper sidelobe level to 15dB, the co-channel interference level is reduced back to appropriate levels. Designs providing greater than 23dB upper- sidelobe levels can solve extreme siting problems. These antennas provide at least 10dB of improvement in co-channel interference over a standard antenna.

You will find a variety of cause-and-effect relationships between the performance parameters. When reducing the first upper sidelobe, the antenna's pattern also lowers some of the other upper sidelobes, which can further reduce the potential for interference. They also naturally exhibit lower intensity beyond the first upper sidelobe.

When the aperture distribution is modified to suppress the upper sidelobes, the pattern's main beam slightly broadens, which results in a minor gain reduction. However, if it is properly designed and manufacturing processes are controlled, an USS antenna will exhibit negligible gain reduction. Generally, the slight gain reduction is offset greatly by the mitigating effects of lower sidelobes.

Beyond USS Although USS can improve your interference problems greatly, you should look for other specific qualities in base-station antennas to further reduce the damaging effects of interference. There is a wealth of information that you must consider in order to maximize initial network-design performance.

But maintaining performance does not end with the initial design. You must continuously optimize the network under live operation to combat usage-pattern shifting, scatter-center movement and urbanization escalation because these factors lead to cell splitting, building donations and base movement. And all of these forces lead to potential interference problems. But with the right mix of base-station antenna products, including USS, you can solve these problems, keep customers happy and, as a result, remain profitable.

In improperly designed antenna can plague the field engineer, wasting valuable resources and adding weeks to a roll-out,maintenance or upgrade schedule. A host of interference issues arises when you begin investigating network performance limitations.

For example, front-to-back ratio (F/B) is important in a low-interference antenna design. Interfering signals like the ones transmitted or received through the upper sidelobes also can seep through the pattern's backlobes. F/B refers to the intensity ratio of the pattern at the peak of the main beam to the intensity in the rear direction. F/B is the measure of an antenna's ability to withstand rearward interference. It is not enough to provide just a deep null in the rearward direction of the pattern, especially if you consider only the primary polarization. A strong antenna product is designed with a maximum allowable backlobe level within an angular range of at least plus/minus 10 degrees about the rearward angle opposite the main beam for any polarization.

Although base-station antennas comprised of log-periodic-type elements typically exhibit higher F/B ratios than traditional low-profile panel antenna arrays, both their rear pattern shapes are corrupted independently when installed on towers, poles and buildings. F/B is a strong function of the environment and exhibits field performance that is seldom the same as what is measured on an antenna test range.

When an antenna is dedicated to a transmitter or equipped for duplex operation at high power levels, it should be designed for low passive intermodulation interference (PIM) generation. PIM is created when the signals from two transmitter channels encounter a poorly designed metal-to-metal contact junction. This junction can exhibit a non-linear electric current conduction process similar to a semiconductor diode. Two transmitter signals encountering a poor junction can form new signals at harmonic frequencies, which are related to the original signal frequencies.

A low PIM design will ensure that any metallic materials that come in contact with one another are similar in electrical composition. The mechanical design also is important in PIM design. Any metal contact that can loosen under environmental stress or become contaminated by other materials can be a source of PIM. It is important to qualify a new design through environmental vibration tests on the antenna before you measure PIM. Furthermore, high transmit power leads to conductor heating in the high electric current path. The variation of thermal loads creates stress in the metallic materials and contacts as well. Therefore, PIM is a conduction process that is a time-dependent phenomena. Be sure your antenna supplier is aware of this phenomenology. Otherwise, the supplier may perform an instantaneous PIM test, which can create misleading results. A reliable PIM test will last for at least five minutes to allow a steady state condition.

High transmit port-to-receive port (TX-RX) isolation also is important for low interference. The transmit port must be highly isolated electrically to coupling of RF signals into the receive ports to avoid burnout or interference effects in the receiver. According to an FCC ruling, antenna products for 800MHz systems must transmit with vertical polarization. This implies that transmit antennas are physically removed from the receive antennas and, if properly located, will provide isolations 40dB or better without faltering.

In other frequency bands, such as PCS, where there is not a mandate to transmit vertically, you can deploy duplexers with the panel antennas. The duplex filters provide sufficient TX-RX isolation for the shared polarization port. When space diversity is implemented, there is sufficient isolation between the physically separated transmit and shared transmit/receive antennas. When polarization diversity is implemented, it is important that the antenna provide sufficient isolation between the two independent polarization ports. The most popular polarization-diversity antenna products use slant 45-degree linear polarization to maximize system diversity gain. The degree of isolation required depends on your frequencyband.>CNWIN Services>TIVoice to Drive Enhanced Services>BYMichael Gulled ge and George Backhaus

Studies show that 20% to 70% of wireless subscribers have indicated an interest in enhanced services. Therefore, the revenue potential is there for the taking. Carriers just need to find ways to turn that interest into usage. Enhanced services such as voice and fax messaging or single-number service allow carriers to provide different network service offerings, drive network usage and generate additional service revenue.

Although subscribers may be clamoring for enhanced services, many of them balk at a product that appears too complex. Ease of use is the key to widespread deployment of enhanced services, and speech is an interface that is easy. Voice-activated services allow carriers to use a product that appeals to a niche market and obtain more revenue from a broader base.

Speech is Natural The first thing we do is teach our offspring to speak -- not press keys -- to get what they want. Speech is a natural way to communicate, and it is hard to regard an interface that most of us learned at our mothers' knees as difficult to master. To use voice-activated services, the subscriber declares what he wants in simple, straightforward phrases. "Call home office." "ET phone home."

First impressions are important, and with voice-activated services, a subscriber's introduction to enhanced services most likely will be a positive one. Enhanced services, such as outdialing from the voice mailbox, drive additional airtime. Subscribers using voice-activated services to check voice mail also can place return calls while still in the voice mailbox.

Subscribers become accustomed to relying on the wireless device and are more receptive to migrating to additional enhanced services, which may or may not require access by a voice interface.

DTMF or touch-tone interfaces originally were developed for simpler messaging activities, such as "To play your messages, press 1." However, market demand for new services has resulted in offerings that can be complex to access through the DTMF keypad. A premium single-number service, for example, can require subscribers to input a series of telephone numbers (home, office, mobile phone, pager) and even require entry of schedules based on time of day and day of week. The market these services appeal to could broaden significantly if, instead of pressing a series of keystrokes, subscribers merely speak.

WIN Advantages One of the keys to voice-activated, enhanced-service offerings is that the technology is put on a service node-type of architecture rather than acting as a standalone point solution. Enhanced services can be deployed in an incremental fashion as a service node, and carriers can leverage capital cost over multiple revenue streams.

Network-based services allow carriers to market enhanced services across all subscribers, without coordination among various handset suppliers. Carriers realize advantages by offering their subscribers voice-activated services that are integrated in the wireless network instead of in the handset itself. They can offer subscribers additional enhanced services seamlessly, and subscribers do not have to purchase new phones to get the additional services.

There are limitations to applications available from phone-based technology. Subscribers cannot migrate to applications such as a personal assistant. Also, as the intelligent network evolves, carriers can migrate to more powerful technology without going to each subscriber's handset.

Churn Reduction Voice-activated services become more efficient the more they are used because the recognition improves. The voice application becomes more like a good friend who knows the subscriber's habits.

The personal directory, or VAS database, is an investment of the subscriber's time and effort, but the return on that investment can be high. The customer programs a list of important phone numbers and locations into the personal directory, and the more he uses the service, the greater the reliability. The personal directory is a powerful deterrent to subscriber churn because having made an investment in time and effort, the subscriber is less likely to go elsewhere.

Voice-activated services also unify applications that subscribers currently may view as separate functions. For example, the same personal directory used to place calls can be used to address voice and fax messages. This "unification" opens the path to migration to other enhanced services.

The Ultimate Test Markets for new products and services take time to develop, and consumers can be resistant to change. But if voice-activated services can increase penetration and lead to other integrated services, the return on investment is great for carriers.

The voice mailbox is the cornerstone to enhanced services. Enhanced services facilitate an ongoing relationship with the customer, a relationship that can be cultivated and constantly updated with services that provide value. For wireless carriers, getting their subscribers to rely on their voice mailboxes means they rely on their carriers. Subscribers develop loyalty and are less likely to churn. Voice-activated services are an easy pathway to the profitable voice mailbox.

The catch-phrase is: "Does this application pass the 'grandmother test?'" It is one thing for the programmers and tech-heads to be able to use the service, but if their grandmothers can use it as well, then you have a service that will be accepted in the marketplace. If carriers give grandmothers the opportunity, they'll talk.