Historical Overview of basic RF Concepts

Transcription

1 Historical Overview of basic RF Concepts High Performance Integrated Circuits Design Group RF Communications System-on-Chip International Master Course E.T.S.E.T.B. D.E.E. Electronics Engineering Master Course Departament d Enginyeria Electrònica Universitat Politècnica de Catalunya Contents: 1. Transmitters 2. Wave detectors (receivers) 3. Who invented the radio? 4. Valves, audions 5. Heterodyne receivers 6. Other coherent receivers 7. The invention of the semiconductor transistor 8. The Integrated Circuit Bibliography: 1. The Design of CMOS Radio-Frequency Integrated Circuits, Thomas. H. Lee. 2. The Design and Implementation of Low-Power CMOS Radio Receivers, Derek K. Shaeffer, Thomas H. Lee 3. Famous Electrochemists web page, maintained by Eugenii Katz (now removed from online access) 4. Wikipedia, the free encyclopedia:

2 RADIOWHAT???? 1890: Radiotelegraphy coded information (wave - no wave : digital transmission) 1906: Radio communication audio transmission (analog modulation of a carrier wave) 1959: First Integrated circuit beginning of the microelectronics era 1990: RF-IC: Advanced modulation techniques improvement of silicon technology Typical transmitter in RF-IC: TRANSMITTER Tx 0 Band Filter Power Amplifier Up Conversion Mixer Modulator Amplification? ω 0 Information encoded and modulated in a waveform Single tone, occupies an specific part of the spectrum

3 Spark transmitter: Hertz invented the first transmitter (emitter) when trying to demonstrate Maxwell s theory about the existence of electromagnetic waves. Hertz, ~1888 A capacitor is charged to a high voltage by an induction coil. When the potential across it is sufficiently high to break down the insulation of air in the gap, a spark occurs. Symplified scheme Emitted wave rich in harmonics (broadband). Since the spark has a low resistance (an ohm or two), the spark discharge is equivalent to the closing of an L-C-R circuit. The condenser then discharges through the conducting spark, and the discharge takes the form of a damped oscillation, at a frequency determined by the resonant frequency of the spark transmitter. Spark transmitter: Antenna: an end-loaded dipole was first used (the frequency transmitted was determined by the self resonant frequency of the antenna system: hundreds of MHz, due to antenna size) Sphere capacitors were used Q C = = 4πε o radi V Another spark gap was used as receiver (detector) (reception was indicated by a visible spark: only useful in small ranges otherwise huge powers would be needed and sparks would rise everywhere, a public health problem.) Hertz, ~1888 Scheme of the first spark transmitter

4 Spark transmitter: Hertz did not imagine his invention useful: It's of no use whatsoever" Hertz, ~ Sine-wave transmitter: The problem with the spark transmitter is that the signal generated was broadband. No two simultaneous communications were possible because of interference. Poulsen, 1903 In 1899, Marconi provided coverage of the America s Cup yacht race. It was a big success. In 1901, he tried to repeat the experience. Two other groups were encouraged to transmit information too. No information was received because of interference. Poulsen made the first sine-wave transmitter, with a resonant L-C circuit. (continuous wave, single tone) A Morse key was used to short 1 or 2 turns of the inductor, shifting the frequency and pulling the transmitter in and out of tune with the receiver (digital FM!). 1 ω o = LC Poulsen arc transmitter

5 Sine-wave transmitter: Any LC resonant tank that is used a sine-wave (carrier) generator needs a negative resistance to compensate losses. Poulsen s arc transmitter used glowing DC arc as negative resistance to maintain oscillation. Still the oscillation was not maintained indefinitely, but was stable enough for radiotelegraphy. NMOS resonant VCO Fessenden in 1906 achieved the first audio radio transmission. His main problem was to generate a continuous carrier. This first voice transmission in 1900 used a spark transmitter; a mechanical device switched it times per second. Poulsen arc transmitter Transmitter power: These transmitters had no amplifiers (were not discovered until 1910). How to transmit to larger distances?? Huge powers (up to 1 MW!!) and huge antennas (up to 150 m!!) were needed to transmit at large distances. Around 1900, Marconi noticed the importance to ground one of the terminals of a monopole antenna. This allowed him the first transatlantic communication in December 12th, Antenna used for the first AM transmission

6 RECEIVER Coherent detection: incoming signal is demodulated by multiplying with a tone which frequency matches that of the transmitted carrier (heterodyne receivers) Easy AM uncoherent detection: rectify & filter Received waveform Rectification and filtering DC blocking Images from Communication Systems, A. Bruce Carlson, McGraw-Hill Coherer detector: Edouard Branly developed an EM wave detector (although he never used it to receive wireless signal transmission) : a tube filled with metal particles. The large resistance between terminals drops several orders of magnitude when an electromagnetic wave reaches it (the electromagnetic wave sticks (or cohers) together the particles inside the tube). Sir Oliver Lodge, in 1894, demonstrated a receiver circuit based on a coherer, that detected electromagnetic waves. It was the first wireless transmission, but no intention to transmit messages was present, just signaling. Branly, 1892 Lodge, 1894 Lodge coined the term coherer. He improved it by automatically shaking the tube after each reception. Lodge later proposed to use matched antennas for tuning the communication. Branly s coherer

7 Crystal detector: The first crystal detector consisted of a small wire (catwhisker) that point-contacted a crystal surface. The device behaved like a diode (despite randomness in the position of the contact, pressure, etc). Several materials were tried (even silicon in 1907!), but galena (lead sulfite) was the most widely used. A resonant LC circuit tuned the transmitted frequency (carrier), the crystal rectified it, and earphones were used to listen to the demodulated signal. Bose, 1901 Early cristal detector Crystal detector: Still no amplifiers: it is expected that a sufficiently strong audio frequency component is received. Main advantages of the crystal detector are its simplicity and no need of battery. Pickard, 1904 antenna Disadvantages are its poor sensitivity (signal must be strong enough to forward-bias the diode without previous amplification). Also earphone must have high impedance. Last, the LC filter provides poor selectivity. headset The detector was later used as the simplest AM receiver. Variable C was used to tune the desired radio channel.

8 BUT WHO INVENTED THE RADIO? Branly, 1892 Marconi, 1895 Popov, 1895 Hertz, ~1888 Fessenden, 1906 Lodge, 1894 Bose, 1901 First radio communication: Popov demonstrated in 1895 an equipment (antenna, coherer) which detected the radiation originated by lightning (he also wrote that could be used for communicating messages) In 1896 he used his equipment to communicate Morse signal between buildings. In subsequent years he improved his radio communication system. Marconi made his first wireless transmission in From the beginning, he had clear idea of using his system for messaging, and had clear idea of its commercial use. Marconi s first experiment s were exactly like Popov s. But Marconi made headlines, told the press, patented his inventions (1897) and created his own communications company (in 1897, which eventually has continuity up to today s Marconi plc). Although Popov and Lodge demonstrated the same physical experiments, they always were behind Marconi s application view. Popov, 1895 Marconi, 1895 Popov s first radio

9 Better receivers: The basic coherer detector was usually triggered by random electromagnetic events (not signals). Moreover it had no tuning capability (not much of a problem with spark transmitters ). Marconi, 1901 Marconi patented (Patent #7777, 1890) a transmitting and receiving system that used filtering and matched antenna lengths to somehow tune the communication. He achieved the first transatlantic communication and commercialized his system. In 1909 his radio system saved 1700 lives when two ferries collided and one of threm sanked off US coast He received the Nobel Prize in 1909 Marconi s transmitter and receiver First voice transmission via radio: Reginald A. Fessenden was the first to experience transmission of voice instead of Morse (dots and dashes) code. (radiotelephony vs. radiotelegraphy) He achieved the first voice transmission in He used a carbon microphone in series with the antenna to modulate a carrier (If the waves can be sent at a high frequency, it is be possible to hear only the variations due to the human voice ). The carrier was generated with a spark transmitter, switched times per second. In Christmas Eve 1906, he used a high-power alternator to generate a carrier and broadcast the first AM emission. (Radiotelegraphists in ships were warned to listen, and they heard a human voice, together with Christmas carols and Haendel music). Fessenden, 1906

10 BETTER DEVICES: VALVES (vacuum tubes) Vacuum tube or valve: John A. Fleming invented the two-terminal vacuum tube, with rectifying properties. Fleming, 1904 It consisted of a bulb with a filament and a plate, each connected to different terminals. Current flowed in one direction (the filament emits electrons when hot, which are collected by the plate), but not in the opposite direction. This effect had already been observed by the inventor of the bulb, T.A. Edison, but gave it no importance. This Fleming valve was used to detect radio signal, but it was little sensitive and consumed much power. Fleming valve

11 Vacuum tube or valve: Lee de Forest had the occurrence to add a grid between the filament and the plate. De Forest, 1906 He called his invention audion (later known as triode), and patented it as detector. He (and others) did not discover its amplifying properties until some years later. The grid current controls the filament to plate current (similar to base terminal in a transistor), thus providing signal amplification. De Forest audion This was the first ever amplifying device, and opened the door to receivers with improved sensitivity. Audion amplifier: Edwin H. Armstrong in 1914 was the first to correctly explain the audion s behavior, and filed the simplest amplifying detector in Armstrong, 1914 It used a galena crystal as detector, and the audion as amplifier. Signals from overseas could be easily detected.

12 HETERODYNE RECEPTION Typical receiver in RF-IC: Rx 0 Band Filter Low Noise Amplifier Down Conversion Mixer Demodulator ω? 0 Assumes narrowband emission Allows multiple transmissions Inherent tuning through ω 0 Heterodyne receiver: Fessenden patented in 1902 the heterodyne receiver (to heterodyne=to mix different signals). Fessenden, 1902 The local oscillator was introduced for the first time. It allowed tuning to the received carrier. Sensitivity was also increased: the oscillator signal was strong enough to switch the diode on (despite weak incoming signals). A significant problem (in later years) was that the oscillator tone was also radiated to the antenna. In subsequent receivers, isolation between the oscillator and the antenna became critical. The heterodyne receiver used frequency conversion for the first time (although this concept was still unrecognized by the inventors).

13 Heterodyne receiver: Ex: 1 MHz Received signal and local oscillator signal were just added. Ex: 1,001 MHz antenna L.O. headset Envelope: 1 KHz The local oscillator was tuned to get a comfortable beat audio frequency 1 KHz ( Superheterodyne receiver: Audions gave little amplification above 1 MHz, thus they could not be used to receive signals above that limit. Following Fessenden s heterodyne concept, he proposed a two-stage receiver. The input RF signal was translated to a intermediate frequency (IF), which could be easily amplified and demodulated. The amplifier and detector work at a fixed IF frequency, while the only tuning happens at the LO. This allows using the same receiver for many RF signals. Armstrong, 1918 This superheterodyne receiver was the first mass-produced AM radio by RCA, and is still the basic receiver architecture used today. Scheme of early superheterodyne receiver

14 Homodyne receiver: The homodyne receiver can be understood as a superheterodyne where the LO frequency equals the RF input, thus providing a zero-if and no need of detection (demodulator). Colebrook first observed this effect, by using a regenerative receiver in which output was overcoupled to the input, thus an oscillation was produced. When oscillation frequency matched the input frequency, no detection was needed. Colebrook, 1924 De Bellescize 1930 Synchronization between input and oscillation was critical. De Bellescize introduced a circuit that guaranteed synchronization by detecting the difference frequency and correcting the LO according to its value. This is the principle of the phase-locked loop, and De Bellescize is considered the inventor of the PLL. AM RECEPTION: NOT A SINGLE SOLUTION

15 Regenerative receiver: The regenerative receiver used a single audion to detect and amplify. Armstrong, 1915 The grid-filament set act like a Fleming valve detector. The plate delivers an amplified replica of the rectified grid signal. The amplified signal was coupled to the input with a RF transformer, to provide positive feedback and thus further amplify the input signal. Regenerative receiver: Armstrong introduced series inductances (input, plate) that acted as impedance matching networks (thus increasing the available signal). Armstrong, 1915 The use of matching inductances and selective feedback provided high Q, thus narrow bandwidth and high selectivity. A number of innovative versions of the circuit were later made, like a differential one for noise rejection. Also, by increasing the coupling between the output and input, the circuit becomes an oscillator.

16 Superregenerative receiver: Amplifier that can achieve gains as high as with the minimum number of components (still used today in cheap circuits) Armstrong, 1922 It is based on an unstable regenerative amplifier. The circuit is initialized periodically, thus it never saturates. After a given period of time, the amplitude will be proportional to the initial condition (the input signal). The resulting output is thus a series of oscillation bursts whose amplitude is the input s amplitude amplified. This can be detected by a simple AM amplifier. Note the circuit amplifies samples of the input, with a frequency higher than the signal bandwidth and lower than the RF carrier. FREQUENCY MODULATION Amplitude modulation (AM) had the problem of being very sensitive to noise and interference (particularly by meteorological phenomena). In AM, the information is contained in the amplitude, while the carrier frequency is constant. He decided that the only solution to this problem was to place information in variations of the carrier frequency (frequency modulation, FM), while keeping the amplitude constant. Armstrong, 1933 First FM modulator

17 GOOD BYE VALVES, WELCOME SEMICONDUCTORS A BJT mixer A practical amplifier circuit First PN junction diode: World War II pushed the need of high-frequency amplifiers for radar applications. Vacuum tubes could not achieve high frequency In 1939, Russell Ohl, a (forgotten) researcher at Bell Labs, tried with the (also forgotten, at that time) catwhisker detector invented by Bose in After experimenting with different materials, he (accidentally) came up to discover a semiconductor diode (PN junction). Germanium diodes were immediately put into production. Oleg Losev had (independently, and also forgotten) developed solid state amplifiers working at 5 MHz already in Ohl 1940

18 First semiconductor transistor: After the War, Shockley, Bardeen and Brattain pursued the fabrication of a semiconductor triode. This was obtained in 1947, by cutting the tip of a plastic triangle covered with a gold plate (thus two close-by contacts were created), which contacted a semiconductor (point-contact transistor) Shockley, Bardeen, Brattain 1947 First transistor (Bell-Labs, 1947) First BJT transistor: In 1950, Shockley created the first real bipolar junction transistor, or BJT. Shockley, 1950 Shockley, Bardeen and Brattain shared the Nobel prize in 1956 Other researchers had previous theoretical descriptions of semiconductor transistors (Lilienfeld, 1928; Heil, 1934), but there is no evidence of their implementation before Also an European group developed the a semiconductor transistor almost simultaneously (Mataré, Welker, 1948)

19 THE FINAL STEP. First Integrated Circuit: The first integrated circuits were manufactured independently by two scientists: Jack Kilby of Texas Instruments (he developed an oscillator), and Robert Noyce from Fairchild Semiconductor (later Intel) Kilby, Noyce 1959 Although Noyce s invention was about 6 months later, a large patent dispute between the two companies started. Jack Kilby was awarded the Nobel Prize in First planar IC, 1961 First Integrated Circuit (Texas Instruments, 1958)

20 Moore s Law: Source: Intel The number of components in an IC doubles every 18 months

21 Source: Intel The number of components 15 nm in an IC doubles every nm months System-on-a-Chip (SoC): 0,35 µm CMOS H. Darabi, et al, A 2.4 GHz CMOS Transceiver for Bluetooth, IEEE ISSC Transilica 0,25 µm CMOS (RF IC) 0,25 µm CMOS Frank Op t Eynde, et al., A Fully-Integrated Single-Chip SoC for Bluetooth, IEEE ISSC 2001

22 System-on-a-Chip (SoC): SoC GPSs 1.8 mm fo=1,57 GHz Sensitivity -130 dbm (19 db below thermal noise) ASSIGNMENTS FOR NEXT WEEK: Download SysCalc from and install it in your personal computer. Navigate the menus and discover its functionalities. Complete the tutorial. Download AppCAD from and install it in your personal computer. Navigate the menus and run the different utilities to discover its functionalities (Can be done in pairs): Search for some wireless communication standard and write in a table the main (for you) standard specs.